On 24 mar, 12:53, Barukh Ziv <barukh....@gmail.com> wrote:
> Guys,
>
> Thank you for the quick reply. I will answer to both questions:
>
> > Are you using ntl_ZZ_p or ntl_ZZ_pE? I have experienced the same type
> > of errors with the latter (due to bad manipulations of
> > ntl_ZZ_pEContext by myself.
>
> Yes, I experience problems while using ZZ_pE as in this case NTL does
> block allocation, and cannot re-allocate a single ZZ_p.
>
> On Mar 24, 11:08 am, Mike Hansen <mhan...@gmail.com> wrote:
>
> > Are you able to post the code that causes this problem? That would
> > help a lot with trying to figure out what's going on.
>
> > --Mike
>
> Mike, unfortunately, the code is too complicated to be presented here,
> but I will try my best to explain what kind of the problem
> I am facing. Basically, the problem occurs once an ZZ_pX was created
> with a certain modulus N, and then assigned from another
> ZZ_pX created with a bigger modulus N' > N. Semantically, consider the
> following algorhtm:
>
> ZZ_pE a, ap;
> // Initialize ap
>
> while (precision < max_precision)
> {
> a = f(ap); // *
> ap = a; // **
>
> precision++;
>
> }
>
> where calculations in f() are done with modulus 2^precision. IMHO,
> both statements (*) and (**) may lead to an error, as they should be
> overwritten with bigger array.
>
> This is essentially a problem stated in the aforementioned SAGE doc
> page on p-adic element. So, I wonder, do you know any good ways to
> overcome this? In particular, is there a way to explicitly re-allocate
> a previously allocated ZZ_pX object?
Yes you can, altought if your code is very low level I am not sure on
the impact.
Consider this:
sage: a = ntl.ZZ_pContext(128)
sage: polynomial = ntl.ZZ_pX([1,2,3,4,5],a)
sage: for i in range(128):
....: a = ntl.ZZ_pContext(a.modulus()*2)
....: f = f.convert_to_modulus(a)
....:
sage: f
[1 2 3 4 5]
sage: f.get_modulus_context()
NTL modulus 43556142965880123323311949751266331066368
This create a new polynomial modulus the right context.However, this
will not work to change context in ZZ_pEX in which you are changing
your defining polynomial (succesive residuals of a p-adic f)
For example,
Suppose f = 15 + x ^2 and you want an algorithm to compute in
(ZZ/2^nZZ)[x] / ( f(x) )
sage: p = ntl.ZZ_pContext(2)
sage: f = [15, 0, 1]
sage: pe = ntl.ZZ_pEContext( ntl.ZZ_pX(f, p) )
sage: poly = ntl.ZZ_pEX([[1], [0], [0,1]], pe)
sage: p2 = ntl.ZZ_pContext(4)
sage: poly = poly.convert_to_modulus(p2)
sage: sage: poly.get_modulus_context()
NTL modulus [1 0 1] (mod 4)
Which is not what we want. Instead, you could add to ntl_ZZ_pEX a
function like (In the following function one should omit the pcontext
input c and use cE.get_pc() instead)
def convert_to_pE(self, ntl_ZZ_pContext_class c,
ntl_ZZ_pEContext_class cE):
"""
Returns a new ntl_ZZ_pE which is the same as self, but
considered modulo a different pE (but the SAME polynomial).
"""
cE.restore_c()
cdef ntl_ZZ_pEX ans = PY_NEW(ntl_ZZ_pEX)
_sig_on
ZZ_pEX_conv_modulus(ans.x, self.x, c.x)
_sig_off
ans.c = cE
return ans
with this function, the previous example would be:
sage: p = ntl.ZZ_pContext(2)
sage: f = [15, 0, 1]
sage: pe = ntl.ZZ_pEContext( ntl.ZZ_pX(f, p) )
sage: poly = ntl.ZZ_pEX([[1], [0], [0,1]], pe)
sage: p2 = ntl.ZZ_pContext(4)
sage: pe2 = ntl.ZZ_pEContext( ntl.ZZ_pX(f, p2) )
sage: poly = poly.convert_to_pE(p2, pe2)
sage: poly.get_modulus_context()
NTL modulus [3 0 1] (mod 4)
Another question: does NTL
> perform an automatic garbage collection in any way?
No idea, sorry
--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
To unsubscribe from this group, send email to
sage-support+unsubscribegooglegroups.com or reply to this email with the words
"REMOVE ME" as the subject.
